1. Field of the Invention
This invention relates to a multi-processor system making use of two or more processors.
2. Description of the Prior Art
There have been provided in the prior art a variety of multi-processor systems making use of a plurality of processors or what are called CPUs. In general, the processors employed in these systems are comparable in their functions or performance.
When it is desired that a variety of software artifices evolved in connection with e.g. an 8-bit CPU computer system be simultaneously used in a 16-bit CPU system having improved functional properties, it is necessary to evolve a multi-processor system with an improved compatibility (or convertibility) between the 8-bit CPU and the 16-bit CPU having widely different functions and above all the address space sizes.
When designing a multi-processor system which employs plural processors with widely differing properties and address space sizes, such as 8-bit and 16-bit CPUs, it is not desirable that the main memory be commonly accessed by the respective processors, because then the main memory capacity is limited by the 8-bit CPU to having a smaller address size so that the excellent properties of the 16-bit CPU are not displayed sufficiently. It is also not desirable to have an independent main memory in each CPU because then the memory utilization efficiency would be lowered.
Memory devices with a memory capacity larger than that corresponding to the address space of an 8-bit CPU system can be controlled by the 8-bit CPU system by resorting to, for example, what is called memory bank commutation or memory banking. However, if these memory devices are simultaneously controlled by the 16-bit CPU, interfacing with the various memory devices will be necessary, thus increasing the system complexity. Moreover, address space allocation must be performed for each memory device, resulting in a correspondingly reduced address space of the 16-bit CPU.